Module thermal field sensing and characteristic analysis method and system

Abstract

The application discloses a module thermal field perception and characteristic analysis method and system, and the method is applied to a data processing device. The method acquires real-time running power of a module, and temperature and heat flow data collected by a perception array attached to a surface of the module; wherein the perception array is divided into core, diffusion and weak interaction hierarchical region according to heat dissipation intensity characteristics, and perception nodes with different performance parameters are arranged in each region to realize non-uniform perception; non-uniform spatial interpolation processing is carried out based on region spatial topology and heat flow density change trend, and a continuous three-dimensional thermal field model of the module is constructed; distribution characteristics in the model are extracted, and core thermal characteristic quantitative indexes including thermal field uniformity, heat dissipation capacity and heat aggregation risk are calculated; and finally, a thermal field dynamic evolution early warning result is output based on the indexes. The application realizes multi-dimensional parameter collaborative perception and three-dimensional reconstruction of a thermal field, and provides objective and accurate quantitative analysis and early warning for safe operation of the module.

Description

Technical Field

[0001] This invention relates to the field of thermal management and monitoring technology for energy storage systems, and in particular to a method and system for module thermal field sensing and characteristic analysis. Background Technology

[0002] As energy storage systems and high-power electronic devices continue to evolve towards higher power and higher energy density, the internal thermal management state of the module directly affects the operational safety and lifecycle reliability of the product. In actual module operation, the distribution, transfer, and evolution of heat within the module is often a highly complex dynamic process, and the heat generation intensity and heat dissipation characteristics of different physical regions of the module typically exhibit significant differences. Therefore, effective state perception and evolution analysis of the module's actual thermal field is not only a crucial step in optimizing system thermal design but also an important foundation for early intervention and dynamic warning of thermal runaway.

[0003] However, traditional module thermal monitoring methods mostly rely on discrete or uniformly distributed single temperature acquisitions, which often fail to comprehensively and accurately reflect the complex heat flow distribution and dynamic transfer patterns within the module. Due to the lack of collaborative sensing of multi-dimensional parameters and the ability to construct continuous spatial thermal fields, existing monitoring methods typically only provide localized static temperature characterizations, making it difficult to accurately and quantitatively assess the overall thermal field distribution differences, actual heat dissipation efficiency, and potential risks of localized heat accumulation within the module. This limitation, to some extent, restricts in-depth analysis of the module's true thermal characteristics and fails to meet the practical needs of modern high-density modules for forward-looking judgment and refined early warning of dynamic thermal field evolution trends. Summary of the Invention

[0004] The purpose of this invention is to provide a module thermal field sensing and characteristic analysis method and system to solve the technical problem that single temperature monitoring is difficult to accurately quantify and assess the module thermal field distribution, heat dissipation capacity and local heat accumulation risk in the prior art.

[0005] To achieve the above objectives, a first aspect of the present invention provides a module thermal field sensing and characteristic analysis method, applied to a data processing device, characterized in that it includes:

[0006] The module acquires real-time operating power and acquires temperature and heat flow data of the module grid cells collected by a digital sensing array deployed on the surface of the module body via a heat-conducting structure. The digital sensing array has non-uniform sensing characteristics, and the module grid cells are pre-divided into core, diffusion, and weak interaction layer regions according to the heat dissipation intensity characteristics of the module. The non-uniform sensing characteristics include at least deploying sensing nodes with different performance parameters in the core, diffusion, and weak interaction layer regions.

[0007] Based on the spatial topological relationship of the multiple hierarchical regions and the heat flux density variation trend reflected by the heat flux data, non-uniform spatial interpolation processing is performed on the temperature data and the heat flux data to construct a continuous three-dimensional thermal field model of the module.

[0008] Extract the temperature distribution characteristics and heat flow distribution characteristics from the three-dimensional thermal field model, and calculate the core thermal characteristic quantitative indicators of the module. The core thermal characteristic quantitative indicators include at least the thermal field uniformity index calculated based on the temperature distribution difference measurement, the heat dissipation capacity index calculated based on the heat flow density, and the heat accumulation risk index calculated by combining the extreme values ​​of temperature and heat flow.

[0009] Based on the aforementioned core thermal characteristic quantification indicators, the system outputs early warning results for the dynamic evolution of the thermal field of the module.

[0010] In one possible implementation, the process of calculating the core thermal characteristic quantification index includes:

[0011] Calculate the standard deviation or variance of the real-time temperature value of each sensing node in the digital sensing array and the average temperature value of all sensing nodes, and construct the thermal field uniformity index based on the ratio of the standard deviation or variance to the average temperature.

[0012] The average value of the real-time heat flow data acquired by the sensing nodes deployed in the diffusion grid unit is calculated, and the ratio of this average value to the real-time operating power of the module is calculated, which is used as the heat dissipation capacity index.

[0013] In one possible implementation, the process of calculating the core thermal characteristic quantification index includes:

[0014] Determine the first ratio of the maximum heat flux data to the average heat flux data of all sensing nodes in the digital sensing array, and the second ratio of the maximum temperature data to the average temperature data.

[0015] The first ratio and the second ratio are multiplied to obtain the heat accumulation risk index used to characterize the tendency of local overheating;

[0016] The process of outputting early warning results of the dynamic evolution of the thermal field for the module based on the core thermal characteristic quantification index includes:

[0017] The standardized temperature and heat flow digital signals after analog-to-digital conversion are transmitted to peripheral devices; and when the heat accumulation risk index is greater than or equal to a preset risk threshold, a local overheating warning signal is generated.

[0018] A second aspect of the present invention provides a module thermal field sensing and characteristic analysis system, including a data processing device and a digital sensing array communicatively connected to the data processing device;

[0019] The digital sensing array is attached to the surface of the module body through a heat-conducting structure and is divided into multiple hierarchical regions according to the heat dissipation intensity characteristics of the module body. The multiple hierarchical regions include core grid units with dense heat flow, diffusion grid units with diffused heat flow, and weak interaction grid units with weak heat flow interaction.

[0020] The digital sensing array includes dual-parameter sensing nodes deployed in the multiple hierarchical regions, and each dual-parameter sensing node is equipped with a temperature sensing element and a heat flow sensing element.

[0021] The data processing device includes a processor and a memory, the memory being used to store computer programs, and the processor executing the computer programs to implement the module thermal field sensing and characteristic analysis method as described in the first aspect above.

[0022] In one possible implementation, to adapt to the heat flow characteristics of each level of region:

[0023] The response speed of the temperature sensing element deployed in the core grid cell is greater than that of the temperature sensing element deployed in the diffusion grid cell, in order to capture transient heat flow accumulation;

[0024] The temperature sensing elements deployed in the weakly interactive grid cells have a higher measurement resolution than those deployed in the diffused grid cells, enabling them to accurately capture boundary heat dissipation anomalies and minute temperature fluctuations.

[0025] In one possible implementation, the temperature sensing element and the heat flow sensing element in the dual-parameter sensing node are arranged side by side in space, and a heat insulation groove is provided between the temperature sensing element and the heat flow sensing element.

[0026] The dual-parameter sensing node is covered with an insulating structure on top and has a heat-conducting structure connected to the module body on the bottom. The combination of lateral heat transfer and heat conduction is blocked by the upper insulation and the middle heat-insulating groove, so as to isolate the interference of the temperature sensing element's own heating on the heat flow sensing element's measurement.

[0027] In one possible implementation, the thermally conductive structure is a gradient thermally conductive bonding structure, which is configured with a stepped thermal conductivity based on the heat dissipation intensity characteristics of the multiple layer regions.

[0028] In one possible implementation, the thermal conductivity of the thermally conductive structure in the region corresponding to the core grid unit is greater than that in the region corresponding to the diffuse grid unit, and the thermal conductivity of the thermally conductive structure in the region corresponding to the diffuse grid unit is greater than that in the region corresponding to the weakly interactive grid unit.

[0029] In one possible implementation, the gradient thermally conductive bonding structure achieves a stepwise change in thermal conductivity by setting thermally conductive silicone with different thicknesses and different intrinsic thermal conductivity in different layer regions.

[0030] In one possible implementation, the temperature sensing element in the core grid cell is an NTC thermistor, the temperature sensing element in the diffused grid cell is a platinum resistance thermometer, and the temperature sensing element in the weakly interactive grid cell is a high-precision digital temperature sensor; each sensing node in the digital sensing array is integrated and arranged on a flexible circuit board.

[0031] The above-described one or more technical solutions in the embodiments of this application have at least one or more of the following technical effects:

[0032] This invention provides a module thermal field sensing and characteristic analysis method. By pre-dividing the module into core, diffusion, and weak interaction hierarchical regions based on the module's heat dissipation intensity characteristics, and strategically deploying dual-parameter sensing nodes with different performance parameters (temperature and heat flux) in each region, this method overcomes the limitations of traditional single and uniformly arranged sensors, achieving close sensing of the module's complex internal features. Simultaneously, this invention combines spatial topological relationships and heat flux density variation trends for non-uniform spatial interpolation, reconstructing discrete sensing data into a continuous three-dimensional thermal field model, thus compensating for the lack of overall thermal field distribution characterization capabilities in existing technologies. Based on this, this invention extracts features from the three-dimensional model and calculates core quantitative indicators such as thermal field uniformity, heat dissipation capacity, and heat accumulation risk, transforming the ambiguous thermal state into multi-dimensional quantitative data. Furthermore, based on these quantitative indicators, it outputs scientific early warning results for the dynamic evolution trend of the module's thermal field, providing an objective and accurate quantitative analysis dimension for the safe operation and thermal design optimization of high-density modules.

[0033] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0034] Figure 1 This is a schematic flowchart of a module thermal field sensing and characteristic analysis method according to an exemplary embodiment;

[0035] Figure 2 This is a schematic diagram of the architecture of a module thermal field sensing and characteristic analysis system according to an exemplary embodiment;

[0036] Figure 3 This is a cross-sectional view of a dual-parameter sensing node in a module thermal field sensing and characteristic analysis system according to an exemplary embodiment.

[0037] Figure 4 This is a schematic diagram of the arrangement of multiple hierarchical regions of a module body in a module thermal field sensing and characteristic analysis system according to an exemplary embodiment.

[0038] Explanation of reference numerals in the attached figures: 100, data processing equipment; 200, digital sensing array; 210, temperature sensing element; 220, heat flow sensing element; 300, heat-conducting structure; 310, gradient heat-conducting bonding structure; 400, module body; 410, core mesh unit; 420, diffusion mesh unit; 430, weakly interactive mesh unit; 440, heat insulation groove; 450, heat insulation structure; 460, flexible circuit board. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0040] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of systems and methods consistent with some aspects of the invention as detailed in the appended claims.

[0041] Figure 1 This is a schematic diagram of a module thermal field sensing and characteristic analysis method according to an exemplary embodiment, such as... Figure 1 As shown, the method is applied to a data processing device, and the method includes:

[0042] In step S100, the real-time operating power of the module is acquired, as well as the temperature and heat flow data of the module's grid cells collected by a digital sensing array deployed on the module's surface via a heat-conducting structure. The digital sensing array exhibits non-uniform sensing characteristics, and the module's grid cells are pre-divided into core, diffusion, and weak-interaction hierarchical regions based on the module's heat dissipation intensity characteristics. The non-uniform sensing characteristics include at least the deployment of sensing nodes with different performance parameters in the core, diffusion, and weak-interaction hierarchical regions. In actual operating conditions, the heat generation and dissipation within the module exhibit significant spatial differences. To accurately capture the true thermal state of the module's underlying structure, a sensing architecture matching its actual heat flow distribution needs to be constructed. Specifically, while acquiring the module's real-time operating power, a digital sensing array is tightly deployed on the module's surface via a heat-conducting structure. The thermally conductive structure can be made of silicone pads with high thermal conductivity, thermally conductive structural adhesive, or phase change materials. Its main function is to fully reduce the contact thermal resistance between the sensing node and the module surface, ensure that the collected temperature data and heat flow data can reflect the physical state of the module body, and reduce data distortion caused by air gaps.

[0043] Furthermore, to optimize system data processing load and hardware cost while ensuring sensing accuracy, the digital sensing array is designed with non-uniform sensing characteristics. Specifically, the module grid cells are not divided using the traditional equal-spacing method, but are pre-divided into three-dimensional regions—core, diffusion, and weak interaction—based on the heat dissipation intensity characteristics of the module under full load or specific operating conditions. The core region typically corresponds to the area of ​​high-heat-generating components with large heat flux and drastic temperature gradient changes within the module, such as the connection point of high-current busbars or the central area of ​​densely stacked battery cells; the diffusion region is adjacent to the periphery of the core region and mainly undertakes the function of heat conduction and dissipation, with strong heat flow directionality; the weak interaction layer region is located at the edge of the module, far from the main heat source and with relatively weak heat convection.

[0044] Based on the aforementioned regional division, the core of the non-uniform sensing characteristic lies in the targeted deployment of sensing nodes with different performance parameters at different levels of regions. Preferably, the differences in performance parameters are mainly reflected in the sensor's measurement accuracy, sampling frequency, and spatial density. Specifically, in the core region, due to the rapid evolution of the thermal state and the high potential risk of heat accumulation, it is preferable to deploy composite sensing nodes with high sampling frequency and high measurement accuracy for both temperature and heat flow parameters to achieve real-time high-frequency capture of severe thermal fluctuations. In the diffusion region, where heat transfer is relatively slow, sensing nodes with medium sampling frequency and accuracy can be deployed to focus on monitoring the path of heat conduction and the attenuation of heat flow density. In the weak interaction level region, due to the large thermal inertia and extremely slow temperature change, basic sensing nodes with low-frequency sampling are deployed with a lower spatial density. This non-uniform array deployment scheme ensures effective monitoring of the module's key thermally sensitive areas while significantly reducing overall data redundancy, communication bandwidth consumption, and computational overhead for subsequent model solving from the underlying hardware architecture.

[0045] In step S200, based on the spatial topological relationships of the multiple hierarchical regions and the heat flux density variation trend reflected by the heat flux data, non-uniform spatial interpolation processing is performed on the temperature data and the heat flux data to construct a continuous three-dimensional thermal field model of the module. After acquiring the temperature data and heat flux data collected by the bottom-level sensing array, since the actual deployed sensors are usually spatially discrete, it is often difficult to comprehensively and accurately characterize the continuous and complex overall heat flux distribution state inside the module based solely on limited local measurement point data. Therefore, it is urgent to transform the discrete observation values ​​into a continuous three-dimensional spatial data field. To solve this problem, this scheme does not adopt the conventional homogeneous interpolation algorithm that relies solely on geometric spatial distance, such as inverse distance weighting, but proposes a non-uniform spatial interpolation processing mechanism driven by physical properties.

[0046] Specifically, the system first establishes the spatial topological relationships among the multiple hierarchical regions. This topological relationship not only encompasses the absolute geometric distances of each sensing node in the three-dimensional coordinate system, but also deeply integrates the conductive connectivity and thermal impedance distribution characteristics of the module's internal physical structure. For example, for two grid cells that are spatially close in a straight line but are physically blocked by insulating materials, their thermodynamic distance in the spatial topological relationship will be set to be relatively large, thereby avoiding erroneous data associations across physical boundaries.

[0047] Based on this physical topology, this scheme innovatively introduces the heat flux density variation trend reflected by the heat flux data, using it as a key correction dimension for non-uniform spatial interpolation. When estimating the temperature and heat flux state of any unsensored target grid cell within the module, the system not only considers the topological distance between the target grid and surrounding known sensing nodes, but also performs dynamic weight compensation by combining the magnitude and conduction direction of the surrounding heat flux vector.

[0048] Specifically, the aforementioned dynamic weight compensation process is preferably implemented by constructing an improved inverse distance weighting algorithm that incorporates a heat flow direction correction coefficient. Let the target mesh cell and known sensing nodes... The topological distance is Traditional interpolation weights are typically 1 ( (This refers to the distance attenuation index). Based on this, a dynamic heat flux correction coefficient is introduced in this scheme. This allows the target mesh cell to acquire the corrected interpolation weights. .

[0049] Furthermore, the heat flux correction coefficient The value is determined by the sensing node. Real-time heat flux vector collected at the location With nodes Spatial direction vector pointing to the target mesh cell The spatial angle relationship between them determines this. In a feasible arithmetic model, the system calculates... and The inner product (i.e.) This is used to quantify the magnitude of attenuation or enhancement.

[0050] When the target mesh cell is located exactly downstream of the local heat flux vector, i.e., the heat flux density change trend points positively toward this region, corresponding to the aforementioned acute angle state where the inner product is greater than zero, the heat flux correction coefficient... When assigned a value greater than 1, the interpolation weight of the upstream high-temperature / high-heat-flux sensing node for the target mesh will be amplified; conversely, if the target mesh is in the reverse or orthogonal blind zone of heat diffusion, corresponding to a state where the inner product is less than or equal to zero, the heat flux correction coefficient... It is assigned a value less than or equal to 1 to reduce the numerical influence of surrounding high-temperature nodes.

[0051] Preferably, the dynamic non-uniform interpolation processing that deeply integrates spatial topological connectivity and heat flow vector field characteristics effectively overcomes the thermal field distortion or excessive smoothing phenomena that are easily caused by traditional pure geometric interpolation at the complex structural boundaries within the module. This process smoothly maps discrete dual-parameter sensing data to the global grid network of the entire module in a manner highly consistent with the underlying thermodynamic conduction physics, thereby accurately constructing a continuous three-dimensional thermal field model of the module. This three-dimensional thermal field model effectively opens up the path from local point monitoring to global volumetric sensing, providing a high-fidelity data foundation for subsequent accurate capture of microscopic signs of local heat accumulation and for conducting dynamic evolution analysis of the macroscopic thermal field.

[0052] In step S300, the temperature distribution features and heat flux distribution features in the three-dimensional thermal field model are extracted, and the core thermal characteristic quantification indicators of the module are calculated. These core thermal characteristic quantification indicators include at least a thermal field uniformity indicator calculated based on temperature distribution difference measurement, a heat dissipation capacity indicator calculated based on heat flux density, and a heat accumulation risk indicator calculated by combining temperature and heat flux extreme values. After constructing a high-fidelity continuous three-dimensional thermal field model of the module through spatial interpolation, the system faces a large amount of three-dimensional spatial node data. To transform these complex thermodynamic scalar and vector fields into decision-making bases that can be directly invoked for safety management, the system enters the feature extraction and indicator quantification stage. Specifically, the data processing equipment deeply analyzes the three-dimensional thermal field model, extracting its temperature distribution features and heat flux distribution features. The temperature distribution features mainly cover scalar information such as global temperature extreme values, local temperature differences, and spatial temperature gradients within the module; while the heat flux distribution features focus on dynamic vector information such as the conduction direction of the heat flux vector, local concentration, and heat flux across specific physical boundaries. By fusing the above multidimensional features, the system calculates the core thermal characteristic quantitative index used to comprehensively characterize the current true thermal state of the module.

[0053] Specifically, the core thermal characteristic quantification index comprises at least three independent and complementary dimensions. The first is the thermal field uniformity index, which is primarily calculated based on a measure of temperature distribution differences. In practice, the system extracts temperature values ​​from grid cells across the entire domain or key areas, and obtains this index by calculating the variance, standard deviation, or volume-weighted root mean square error of the spatial temperature. This thermal field uniformity index objectively quantifies the degree of temperature polarization within the module caused by structural layout or heating differences; its numerical variation effectively reflects the non-uniform aging environment and potential lifespan degradation trends faced by the internal battery cells.

[0054] Secondly, there is the heat dissipation capability index, which is mainly calculated based on heat flux density and focuses on evaluating the dynamic efficiency of the module in dissipating internally accumulated heat to the external environment. Preferably, the system extracts heat flux vector data located at the physical boundary of the module (such as the surface with a thermally conductive structure), performs surface integral on the outward heat flux density, and normalizes it by combining the temperature difference between the module core area and the external cooling medium. This index accurately characterizes the actual heat dissipation rate driven by a unit temperature difference and can be effectively used to monitor physical-level heat dissipation channel obstruction faults such as thermal conductive silicone pad delamination and aging or micro-blockage of the liquid cooling system.

[0055] Furthermore, to overcome the technical bottleneck of the lag in traditional single-temperature alarms, the system calculates a heat accumulation risk index, a high-dimensional safety parameter that combines temperature and heat flow extremes. Specifically, the system not only tracks the margin of local highest temperature nodes approaching preset safety thresholds in real time within a three-dimensional thermal field, but also simultaneously monitors the abnormal rate of change of heat flow density in that local area. When a micro-region exhibits an abnormally high absolute temperature and local heat flow shows an extreme accumulation trend of high heat concentration that cannot be effectively diffused, the heat accumulation risk index calculated by the system through nonlinear weighting will show a significant upward trend. This quantitative analysis mechanism, which integrates static temperature states with dynamic heat flow accumulation trends, can more sensitively and proactively capture hidden danger signals that induce early thermal runaway, such as micro-short circuits in battery cells and local insulation breakdown, thus providing reliable data support for early safety intervention.

[0056] In step S400, based on the core thermal characteristic quantification indicators, an early warning result for the dynamic evolution of the thermal field of the module is output. After continuously calculating and acquiring multi-dimensional core thermal characteristic quantification indicators through data processing equipment, the system then enters the safety control and dynamic decision-making stage. In order to achieve proactive assessment and refined management of the module's thermal state, this solution constructs an objective and forward-looking early warning triggering mechanism based on the aforementioned extracted structured quantitative data. Specifically, the system internally pre-sets a multi-dimensional dynamic safety baseline or threshold matrix that matches different operating conditions of the module. During operation, the data processing equipment extracts the calculated thermal field uniformity indicators, heat dissipation capacity indicators, and heat accumulation risk indicators in real time, and performs high-frequency cross-comparison analysis with the dynamic safety baseline, thereby outputting an early warning result for the dynamic evolution of the thermal field of the module.

[0057] Preferably, to accurately reflect the complex thermodynamic evolution process, the early warning result of the thermal field dynamic evolution is not a simple on / off alarm, but a graded response result formed based on the over-limit characteristics of different core thermal characteristic quantitative indicators. Specifically, when the comparative analysis shows that the thermal field uniformity index deviates from the normal baseline range, while the heat accumulation risk index remains low, the system judges that a slight temperature polarization is occurring inside the module due to differences in heat dissipation or inconsistent internal resistance of components, and then outputs an early warning indicating the deviation in state to indicate potential non-uniform aging risks; when the heat dissipation capacity index is detected to show an irreversible downward trend in continuous evaluation cycles, the system judges that the flow of cooling medium outside the module is obstructed or that the internal heat conduction interface is physically attenuated, and then outputs a medium-level warning indicating obstruction of the heat dissipation path to trigger a proactive maintenance and troubleshooting mechanism.

[0058] More importantly, once the system detects a significant increase in the heat accumulation risk indicator within a short time window, instantly exceeding the preset high-risk threshold, the system will immediately determine that a serious anomaly is occurring inside the module, such as a micro-short circuit or local insulation failure, causing heat to be unable to dissipate. At this time, the system will quickly output the highest-level critical high-risk warning result. This warning result is not only transformed into an intuitive audible and visual alarm signal or a pop-up prompt on the operation and maintenance interface, but also serves as a high-priority control basis to directly link the underlying hardware protection logic, such as instantly issuing a command to cut off the electrical circuit of the abnormal module, or simultaneously activating the matching fire-fighting targeted suppression system. This dynamic early warning mechanism, which deeply integrates the quantification of microscopic thermal characteristics with macroscopic safety response, effectively improves the traditional passive and delayed mode that relies solely on the absolute surface temperature exceeding the limit to trigger an alarm, thus constructing an early-stage active safety defense barrier for high-energy-density modules.

[0059] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0060] In an exemplary embodiment, the process of calculating the core thermal characteristic quantification index includes:

[0061] Calculate the standard deviation or variance of the real-time temperature value of each sensing node in the digital sensing array and the average temperature value of all sensing nodes, and construct the thermal field uniformity index based on the ratio of the standard deviation or variance to the average temperature.

[0062] The average value of the real-time heat flow data acquired by the sensing nodes deployed in the diffusion grid unit is calculated, and the ratio of this average value to the real-time operating power of the module is calculated, which is used as the heat dissipation capacity index.

[0063] After extracting the underlying sensing data of the three-dimensional thermal field model, the system needs to further perform dimensionality reduction and feature extraction using specific statistical dimensions. Regarding the construction of the thermal field uniformity index, the data processing device first acquires the real-time temperature values ​​of each sensing node in the digital sensing array during the current sampling period and simultaneously calculates the average temperature of all sensing nodes. Based on this scheme, the system then calculates the standard deviation or variance of the real-time temperature value of each sensing node relative to the overall average temperature value, thereby characterizing the absolute dispersion of the temperature in each region from the overall temperature center. Subsequently, based on the ratio of the standard deviation or variance to the average temperature, the system constructs a thermal field uniformity index that can effectively eliminate interference from fluctuations in the environmental reference temperature.

[0064] Furthermore, at the level of specific engineering applications and algorithm implementation, considering the computational power limitations of the underlying management chips in energy storage systems, such as the microcontrollers of battery management systems, to avoid excessive computational overhead caused by square root or high-frequency exponentiation operations, it is preferable to simplify the standard deviation or variance calculations mentioned above by using the statistically equivalent mean of absolute deviations. Based on this, using the latest optimized calculation model, the specific calculation formula for the thermal field uniformity index is as follows:

[0065]

[0066] in, This is an indicator of thermal field uniformity. The real-time temperature values ​​of each sensing node; The average temperature of all sensing nodes; This represents the total number of sensing nodes. Under this calculation logic, the numerator of the formula calculates the sum of the absolute differences between the temperature of each node and the average temperature, and then divides it by... This effectively characterizes the temperature variance dispersion characteristics in the aforementioned scheme. Further normalization is performed by dividing it by the average temperature $T$, and then using the overall baseline value... Subtract them. The result is... The closer the value is This means that the smaller the temperature variance or deviation in each area inside, the better the thermal uniformity and cell consistency of the module.

[0067] On the other hand, when calculating the heat dissipation capacity index, the system precisely focuses its monitoring on the diffusion grid cells that perform the core heat transfer function. Specifically, the system extracts real-time heat flow data acquired by the sensing nodes deployed in the diffusion grid cells and calculates the average value of these boundary heat conduction data. Subsequently, the system obtains the real-time operating power of the module at the current moment and calculates the ratio of the average value of the aforementioned real-time heat flow data to the real-time operating power. The physical meaning of this ratio is that it directly characterizes the dynamic response efficiency of the module in dissipating heat under a unit operating load.

[0068] Similarly, when transforming the above scheme into a practical algorithm model, since the number of sensing nodes $m$ is fixed as a constant after the physical structure is finalized, there is a substantially linear and proportional equivalence between the average value of the real-time heat flow data and the sum of the heat flow data. To more intuitively characterize the overall intensity of heat dissipation from the module to the outside, based on the above ratio calculation, it is preferable to directly use the ratio of the sum of the heat flow data to the operating power as the final evaluation parameter. Based on this, the specific calculation formula for the heat dissipation capacity index is as follows:

[0069]

[0070] in, This is an indicator of heat dissipation capacity; Real-time heat flow data, i.e. heat dissipation intensity, is obtained from the sensing nodes of each diffusion grid unit; The number of nodes is sensed for in the diffused grid cell; This represents the real-time operating power of the module. This formula simplifies the complex average conversion to a characteristic parameter characterizing the heat dissipation efficiency per unit power by directly comparing the total output of boundary heat flux with the overall electrical power input of the module. A larger value for this quantitative parameter indicates a smoother heat dissipation path for the module; if a decreasing trend in this value is observed under the same power baseline, it can serve as a precise quantitative basis for the delamination of the underlying heat-conducting medium or the attenuation of the cooling circuit.

[0071] In an exemplary embodiment, the process of calculating the core thermal characteristic quantification index includes:

[0072] Determine the first ratio of the maximum heat flux data to the average heat flux data of all sensing nodes in the digital sensing array, and the second ratio of the maximum temperature data to the average temperature data.

[0073] The first ratio and the second ratio are multiplied to obtain the heat accumulation risk index used to characterize the tendency of local overheating;

[0074] The process of outputting early warning results of the dynamic evolution of the thermal field for the module based on the core thermal characteristic quantification index includes:

[0075] The standardized temperature and heat flow digital signals after analog-to-digital conversion are transmitted to peripheral devices; and when the heat accumulation risk index is greater than or equal to a preset risk threshold, a local overheating warning signal is generated.

[0076] To overcome the significant lag in traditional single absolute temperature alarms and achieve proactive detection of extreme thermal runaway events, this solution extracts extreme and mean features from a three-dimensional thermal field model to calculate a heat accumulation risk index. Specifically, after acquiring global sensing data, the data processing device determines the maximum and average heat flux data, as well as the maximum and average temperature data, for all sensing nodes in the digital sensing array. Based on the extracted feature parameters, the system multiplies the quotient of the maximum and average heat flux data (i.e., the first ratio) with the maximum temperature difference deviation rate (i.e., the second ratio) to obtain the heat accumulation risk index characterizing local overheating tendencies.

[0077] Specifically, the calculation formula for the heat accumulation risk index is as follows:

[0078]

[0079] in, As an indicator of thermal agglomeration risk; The maximum heat flux data acquired for all sensing nodes of the module; Average heat flux data acquired for all sensing nodes; The maximum temperature data acquired by all sensing nodes of the module; Let be the average temperature of all sensing nodes. In this calculation model, the first multiplier intuitively reflects the convergence rate of heat generation in a local area; while the second multiplier innovatively introduces... This serves as a benchmark constant for the evolution of local temperature differences. The physical meaning of this specific formula is that when the local maximum temperature is higher than the overall average temperature and approaches or exceeds it... At that time, the second multiplier will approach or exceed If at this time, the region is simultaneously accompanied by an abnormal convergence of local heat flow (the first multiplier is greater than 1). Multiplying the two will increase the risk index of thermal agglomeration. Nonlinear amplification occurs. This calculation, which combines a fixed temperature difference reference with a dynamic heat flow ratio, can effectively eliminate reasonable temperature difference interference allowed under normal charging and discharging conditions of the module, and accurately locate the real abnormal hot spots.

[0080] After continuously calculating and acquiring the aforementioned core thermal characteristic quantification indicators, the system then executes a process of outputting a dynamic thermal field evolution early warning result for the module. Specifically, this output process includes two parallel mechanisms: underlying data transmission and upper-level logic early warning. On one hand, the system's underlying digital signal calibration and transmission module transmits the original collected and converted (analog-to-digital) standardized temperature and heat flow digital signals in real time to peripheral devices, such as the main control battery management system of the energy storage system or the upper-level operation and maintenance cloud platform, to provide high-precision and unbiased underlying thermal field data support for third-party application systems. On the other hand, the system's internal safety early warning logic unit monitors the calculated heat accumulation risk indicators in real time and compares them with internally preset risk thresholds (preferably numerical values). The system performs comparative analysis. When the system determines that the heat accumulation risk index is greater than or equal to the preset risk threshold, it indicates that there is a high probability of local heat accumulation within the module that cannot be effectively dissipated, and promptly generates a local overheating warning signal. This signal can directly trigger underlying hardware protection actions such as electrical disconnection, thereby constructing an active safety defense barrier when thermal runaway is in its nascent stage.

[0081] In an exemplary embodiment, please refer to Figures 2 to 4 The present invention also provides a module thermal field sensing and characteristic analysis system, including a data processing device and a digital sensing array communicatively connected to the data processing device;

[0082] The digital sensing array is attached to the surface of the module body through a heat-conducting structure and is divided into multiple hierarchical regions according to the heat dissipation intensity characteristics of the module body. The multiple hierarchical regions include core grid units with dense heat flow, diffusion grid units with diffused heat flow, and weak interaction grid units with weak heat flow interaction.

[0083] The digital sensing array includes dual-parameter sensing nodes deployed in the multiple hierarchical regions, and each dual-parameter sensing node is equipped with a temperature sensing element and a heat flow sensing element.

[0084] The data processing device includes a processor and a memory. The memory is used to store computer programs. When the processor executes the computer programs, it implements the module thermal field sensing and characteristic analysis method as described in the above embodiments.

[0085] Based on the core technical concepts constructed by the aforementioned methodological process, this embodiment further provides a module thermal field sensing and characteristic analysis system that is highly coordinated with the method from the perspective of physical hardware and system architecture. Specifically, the system mainly consists of a digital sensing array responsible for the underlying data acquisition and a data processing device responsible for the upper-level logic operations in terms of hardware architecture. The two establish a stable communication connection through an industrial-grade communication bus or a high-speed data interface to ensure that the physical electrical signals acquired by the underlying array can be transmitted to the data processing center in real time, with low latency and without loss.

[0086] At the underlying data acquisition end, the digital sensing array is tightly deployed on the surface of the module body through a specific thermally conductive structure, such as silicone pads or thermally conductive structural adhesives with different thermal conductivity coefficients. To effectively sense the complex heat flow field inside the module, the array abandons the traditional uniform distribution pattern in its spatial layout. Instead, it strictly divides the monitoring space into multiple hierarchical regions with independent thermal properties based on the heat dissipation intensity distribution characteristics of the module body under actual operating conditions. Specifically, these hierarchical regions explicitly include: core grid units corresponding to high-heat-generating sources with highly concentrated internal heat flow; diffusion grid units surrounding the core area, responsible for the main outward conduction and dissipation of heat; and weakly interactive grid units located at the physical edges of the module or in airflow dead zones, where heat flow interaction is relatively weak.

[0087] To address the differentiated thermodynamic characteristics of the aforementioned regions at different levels, the digital sensing array specifically deploys highly integrated dual-parameter sensing nodes. Specifically, each dual-parameter sensing node is physically equipped with both an independent temperature sensing element and a heat flow sensing element. Preferably, to ensure the accuracy of dual-parameter data acquisition, the temperature sensing element and the heat flow sensing element are placed side-by-side inside the node. An insulation layer is provided above the physical structure to block stray interference from external airflow, while the bottom of the node achieves good thermal coupling with the module body through a heat-conducting structure. This effectively reduces cross-interference from the temperature sensing element's own heating on heat flow measurement, ensuring high accuracy of the underlying input data.

[0088] At the data processing and decision-making end, the data processing device serves as the digital brain of the entire system, and its internal hardware architecture includes at least a processor and a memory. The memory typically uses non-volatile storage media, such as flash memory or solid-state drives, to securely and stably store pre-compiled underlying operating systems, system-level configuration parameters, and core computer programs. The processor, as the execution unit, possesses powerful floating-point arithmetic and concurrent processing capabilities. When the system powers on and the processor activates and executes the computer programs stored in the memory, it can completely and smoothly invoke the aforementioned algorithm models, thereby realizing the module thermal field perception and characteristic analysis method detailed in the above embodiments. Specifically, this encompasses everything from non-uniform spatial interpolation processing of multi-source heterogeneous data and continuous reconstruction of three-dimensional thermal field models to the extraction and calculation of three core thermal characteristic quantification indicators: thermal field uniformity, heat dissipation capacity, and heat accumulation risk. Ultimately, it completes high-dimensional cross-judgment and graded early warning output of the dynamic evolution trend of the thermal field. This system effectively constructs a proactive safety protection system that integrates hardware and software, with a closed-loop linkage between underlying physical perception and upper-level digital diagnosis.

[0089] In an exemplary embodiment, to adapt to the heat flow characteristics of each layer region:

[0090] The response speed of the temperature sensing element deployed in the core grid cell is greater than that of the temperature sensing element deployed in the diffusion grid cell, in order to capture transient heat flow accumulation;

[0091] The temperature sensing elements deployed in the weakly interactive grid cells have a higher measurement resolution than those deployed in the diffused grid cells, enabling them to accurately capture boundary heat dissipation anomalies and minute temperature fluctuations.

[0092] To effectively monitor the complex thermal field within the module, this embodiment strictly adheres to a differentiated design principle in the selection and configuration of the underlying hardware, matching the thermal flow characteristics of each layer and region. Since different regions within the module exhibit significant differences in heat generation mechanisms and heat dissipation rates, using globally uniform sensors often leads to the omission of key thermal features or unnecessary waste of hardware resources. Therefore, the system has refined the parameters of the temperature sensing elements deployed within the core grid unit, diffused grid unit, and weakly interactive grid unit, taking into account their unique thermal properties.

[0093] Specifically, the core grid unit typically corresponds to the main heat sources within the module, such as the battery cell tabs and high-power busbars. Its thermal characteristics include high heat flux density and rapid temperature rise under abnormal operating conditions. To accommodate this characteristic, the system preferably configures the response speed of the temperature sensing elements deployed in the core grid unit to be significantly greater than that of the temperature sensing elements deployed in the diffusion grid unit. In practical engineering implementation, this is typically achieved by selecting sensing devices with a small thermal time constant, such as miniature packaged fast-response NTC thermistors. This high-response configuration ensures that the system can capture steep temperature surge curves with a low hysteresis time when facing transient heat accumulation caused by high-rate charging / discharging or micro-short circuits in the battery cells, thus providing timely underlying data support for very early warning of high-risk heat accumulation.

[0094] Meanwhile, the weakly interactive mesh units are located at the physical edges or thermal convection dead zones of the module, exhibiting thermal characteristics of sparse heat flow, extremely slow temperature changes, and very slight amplitude. To prevent minute temperature anomalies from being masked by sensor quantization errors, the system preferably configures the measurement resolution of the temperature sensing elements deployed in the weakly interactive mesh units to be higher than that of the temperature sensing elements deployed in the diffused mesh units. By employing ultra-high sensitivity temperature sensing elements, the system can accurately capture extremely small temperature drifts within this region. This high-resolution monitoring capability plays a crucial role in identifying potential slow heat dissipation anomalies at the module boundaries, such as localized insulation material failure or micro-blockage of edge cooling channels, thereby ensuring the data integrity and high fidelity of the global three-dimensional thermal field modeling at boundary conditions.

[0095] In summary, this differentiated hardware deployment strategy, which uses the steady-state monitoring accuracy of the diffused grid cells as a benchmark and demands high response speed from the core area and high measurement resolution from the weak interaction area, not only adapts well to the actual thermodynamic evolution of the module, but also achieves a reasonable configuration of system hardware cost and data processing bandwidth while ensuring full-dimensional thermal field perception capability.

[0096] In an exemplary embodiment, please refer to Figure 3 The temperature sensing element and the heat flow sensing element in the dual-parameter sensing node are arranged side by side in space, and a heat insulation groove is provided between the temperature sensing element and the heat flow sensing element.

[0097] The dual-parameter sensing node is covered with an insulating structure on top and has a heat-conducting structure connected to the module body on the bottom. The combination of lateral heat transfer and heat conduction is blocked by the upper insulation and the middle heat-insulating groove, so as to isolate the interference of the temperature sensing element's own heating on the heat flow sensing element's measurement.

[0098] In the digital sensing of the module's thermal field, the co-source acquisition of temperature and heat flow signals is fundamental to constructing a three-dimensional thermal field model. However, as is known to those skilled in the art, temperature sensing elements (such as NTC thermistors or PT1000 platinum resistance thermometers) require a weak measurement current during operation, which typically generates a small amount of Joule heat, i.e., a self-heating effect. If a highly sensitive heat flow sensing element is placed close to it, the heat generated by this temperature sensing element itself can easily form stray heat flow through lateral conduction, thus severely interfering with the measurement accuracy of the heat flow sensing element. To effectively reduce this physical-level signal crosstalk, this embodiment employs a precise structural design for the internal packaging and spatial layout of the dual-parameter sensing node.

[0099] Specifically, within the dual-parameter sensing node, the temperature sensing element and the heat flow sensing element are not arranged in a traditional stacked configuration, but rather placed side-by-side. Between the two side-by-side, a physical insulation groove is specifically designed. This insulation groove effectively cuts off the continuous solid heat transfer medium path between them, thereby significantly increasing the lateral thermal resistance and effectively blocking the lateral heat transfer from the temperature element's self-generated heat to the heat flow element.

[0100] Furthermore, around the aforementioned parallel sensing elements, an asymmetric thermal conductivity and insulation system is constructed at the node in the direction perpendicular to the heat flow. Specifically, the upper part of the dual-parameter sensing node, i.e., the side away from the module's heating surface, is entirely covered with an insulation structure. This insulation structure effectively shields the temperature fluctuations caused by external environmental airflow convection or stray radiation from the module, forcing the sensed heat flow to be unidirectional. Simultaneously, below the node, i.e., on the side in contact with the module body, a thermally conductive structure with excellent thermal conductivity is provided. This thermally conductive structure ensures that both the temperature sensing element and the heat flow sensing element can achieve low thermal resistance thermal coupling with the module body, accurately and sensitively acquiring the original thermal state of the module surface. In summary, this embodiment cleverly constructs a closed and directional heat dissipation channel through a combination of upper insulation to block external environmental interference, a middle insulation groove to physically cut off the internal lateral heat transfer path, and lower efficient thermal conductivity to fit the measured source. This structure isolates the temperature sensing element from cross-interference with the heat flow sensing element's measurement caused by its own heating, achieving high-accuracy acquisition of dual-parameter data within a limited physical space.

[0101] Preferably, the thermally conductive structure is a gradient thermally conductive bonding structure, which is configured with a stepped thermal conductivity based on the heat dissipation intensity characteristics of the multiple layered regions. In the module thermal field sensing system, the quality of thermal coupling between the sensing node and the module body directly determines the accuracy of the underlying data acquisition. To eliminate contact thermal resistance while better adapting to the highly non-uniform heat flow distribution characteristics inside the module, this embodiment preferably designs the thermally conductive structure as a gradient thermally conductive bonding structure. Specifically, traditional homogeneous thermally conductive materials often struggle to simultaneously meet the rapid response requirements of high heat flow regions and the material property matching of low heat flow regions. However, the gradient thermally conductive bonding structure in this solution is configured with a stepped thermal conductivity in spatial distribution, strictly based on the actual heat dissipation intensity characteristics of each of the multiple layered regions. This design ensures that the temperature and heat flow measurement components in different regions can achieve good thermal property bonding with the object being measured below.

[0102] Specifically, the stepped thermal conductivity is clearly defined in the selection of physical materials to address the differences in heat flux density exhibited by different grid cells. Preferably, for the core grid cell with the densest heat flux and intense transient heat generation, the thermally conductive structure below it is made of a high thermal conductivity material, with a thermal conductivity greater than or equal to 2.5 W / (m·K); for the diffusion grid cell, which undertakes the main task of heat conduction outward, the thermally conductive structure below it is made of a medium thermal conductivity material, with a thermal conductivity greater than or equal to 2.0 W / (m·K); and for the weakly interactive grid cell, which has the least heat flux interaction and extremely gradual temperature change, the thermally conductive structure below it is made of a basic thermally conductive material, with a thermal conductivity greater than or equal to 1.5 W / (m·K). Through the above-mentioned three-level stepped thermal conductivity configuration (high, medium, and low), the system accurately opens up efficient heat diversion channels in the key core area, ensuring high fidelity data from the high-response probe, while also better matching the physical reality of slow heat dissipation in the edge area.

[0103] Furthermore, in addition to configuring a stepped thermal conductivity at the material level, the gradient thermal bonding structure further incorporates the actual three-dimensional shape of the module body. Based on the different physical assembly gaps and stress buffering requirements of each grid region, thermally conductive sheets of varying thicknesses are selected and configured, such as gradient thermally conductive silicone of different specifications. In actual assembly, the stepped changes in thermal conductivity and the adaptive adjustment of spatial thickness complement each other, jointly constructing a gradient thermal network in three-dimensional space. This network ensures a tight physical bond between the dual-parameter sensing nodes and the module surface under complex operating conditions, avoiding measurement deviations caused by tiny air gaps. It also achieves a high degree of synchronization between the sensing array and the module body's heat dissipation characteristics at the underlying thermodynamic conduction level, thus laying a solid and reliable physical foundation for subsequent high-precision continuous three-dimensional thermal field modeling.

[0104] Furthermore, the thermal conductivity of the thermally conductive structure in the region corresponding to the core grid unit is greater than that in the region corresponding to the diffuse grid unit, and the thermal conductivity of the thermally conductive structure in the region corresponding to the diffuse grid unit is greater than that in the region corresponding to the weakly interactive grid unit. Having clarified the basic architecture of the gradient thermally conductive bonding structure, this embodiment provides a more rigorous physical definition of the specific numerical relationships of the thermal conductivity in each region. Specifically, the system is configured such that the thermal conductivity of the thermally conductive structure in the region corresponding to the core grid unit is greater than that in the region corresponding to the diffuse grid unit, and the thermal conductivity of the thermally conductive structure in the region corresponding to the diffuse grid unit is greater than that in the region corresponding to the weakly interactive grid unit.

[0105] This strictly decreasing thermal conductivity configuration strategy perfectly aligns with the spatial heterogeneity of heat flow conduction dynamics within the module. Specifically, since the core grid unit directly faces the main heat source with intense transient heat generation and high heat flux, configuring its corresponding region with the highest global thermal conductivity significantly reduces the contact thermal resistance between the sensing node and the measured heat source. This design ensures that high-density transient heat flow can be conducted to the dual-parameter sensing node with low hysteresis, thereby providing high response sensitivity for the underlying hardware to keenly detect localized high-risk heat accumulation.

[0106] Based on this, since the diffusion grid unit mainly reflects the steady-state or quasi-steady-state conduction process of heat diffusion from the center to the periphery, the system is configured with a thermal conductivity at an intermediate level for its corresponding region. This moderate thermal conductivity design is particularly crucial, as it ensures the accurate transfer of temperature gradients and heat flow vectors while effectively preventing the thermally conductive structure itself from becoming an additional heat sink due to excessive thermal conductivity, i.e., excessively absorbing and altering the original heat flow direction, thereby preventing distortion of the natural heat flow distribution morphology in this region.

[0107] Furthermore, for the weakly interactive mesh cells where thermal interaction is minimal and temperature changes are relatively gradual, the system configures the corresponding region with the lowest global thermal conductivity. This targeted design effectively prevents the bonding material at the bottom of the sensing node from forming abnormally rapid heat transfer channels locally, avoiding the unreasonable conduction of heat from the core or diffusion region to the edge mesh. By using a relatively low thermal conductivity, the system better reflects the original slight temperature fluctuations and thermal boundary characteristics of the weakly interactive region. In summary, this thermal conductivity configuration, which strictly follows the spatial decreasing law of heat dissipation intensity, ensures high-fidelity mapping of the sensing array to complex three-dimensional thermal fields from the underlying physical material structure.

[0108] Furthermore, the gradient thermally conductive bonding structure achieves a step-like change in thermal conductivity by setting thermally conductive silicone with different thicknesses and intrinsic thermal conductivity in different layer regions. In a further specific embodiment, in order to effectively implement the aforementioned gradient thermal conductivity design at the physical manufacturing and assembly level, this embodiment limits the specific selection and molding process of the gradient thermally conductive bonding structure. Specifically, heat flow conduction inside the module is a complex process constrained by multiple physical quantities. The actual thermal resistance at the heat transfer interface depends not only on the heat transfer properties of the material itself, but also on the direct influence of the physical thickness of the interface material. Therefore, this solution does not adopt the conventional approach of changing only a single variable, but rather the gradient thermally conductive bonding structure achieves a comprehensive step-like change in thermal conductivity on a macroscopic scale by simultaneously setting thermally conductive silicone with different thicknesses and intrinsic thermal conductivity in different layer regions.

[0109] Specifically, at the intrinsic property level of the materials, the system selects thermally conductive silicone substrates with different intrinsic thermal conductivity for the core mesh unit, diffusion mesh unit, and weakly interactive mesh unit. As mentioned earlier, silicone with high intrinsic thermal conductivity is matched for high heat flux regions, and silicone with low intrinsic thermal conductivity is matched for low heat flux regions, thereby constructing a differentiated foundation for heat flow guidance at the material's underlying layer.

[0110] Furthermore, in terms of physical geometry, considering the unavoidable structural tolerances of the module body during actual production and assembly, as well as the bonding stress requirements of dual-parameter sensing nodes in different regions, the system further incorporates thermally conductive silicone of varying thicknesses in the aforementioned different levels of regions. It is well known to those skilled in the art that the thermal resistance of a heat transfer interface is directly proportional to the material thickness. By coupling a specific intrinsically thermally conductive material with a precisely calculated thickness, the system can finely adjust the actual interface thermal resistance of each grid cell. For example, in the core grid cell, a thermally conductive silicone with high intrinsic thermal conductivity and a relatively thin thickness is selected, while ensuring sufficient filling of physical assembly gaps and effective reduction of air thermal resistance, to achieve efficient heat penetration and transient response. In weakly interactive grid cells, a thermally conductive silicone with low intrinsic thermal conductivity and an appropriate thickness can be used to form a gentle thermal coupling state, preventing excessive heat absorption interference to the weak temperature field at the edges.

[0111] In summary, this dual-variable adjustment mechanism, which combines different intrinsic thermal conductivity and different thicknesses, not only better adapts to the complex three-dimensional physical structure and assembly tolerances inside the module, but also smoothly and effectively achieves the expected stepwise change in thermal conductivity in terms of actual thermodynamic conduction. This provides a reliable physical interface support for the global sensing array to acquire high-fidelity, low-bias underlying heat source signals.

[0112] In an exemplary embodiment, the temperature sensing element in the core grid unit is an NTC thermistor, the temperature sensing element in the diffused grid unit is a platinum resistance thermometer, and the temperature sensing element in the weakly interactive grid unit is a high-precision digital temperature sensor; each sensing node in the digital sensing array is integrated and arranged on a flexible circuit board. To further implement the aforementioned differentiated sensing strategy based on heat dissipation intensity into the specific underlying hardware architecture, this embodiment clearly defines the type of temperature sensing element in each grid unit and the physical packaging form of the overall array. Specifically, because the thermodynamic monitoring requirements faced by different areas within the module vary significantly, the system abandons the global deployment scheme of a single sensor and instead adopts a multi-source heterogeneous sensor combination matrix.

[0113] Firstly, at the hardware component selection level, the system configures the temperature sensing element in the core grid unit as an NTC (negative temperature coefficient) thermistor. Those skilled in the art know that NTC thermistors possess physical characteristics such as small size, short thermal time constant, and high sensitivity to temperature changes. Deploying them in the area with the highest heat density can effectively leverage their rapid response advantage, thereby quickly capturing transient heat flow accumulation and temperature surges caused by high-rate discharge or micro-short circuits in the battery cell.

[0114] Secondly, the system configures the temperature sensing element in the diffusion grid unit as a platinum resistance thermometer, such as the common PT100 or PT1000. Compared to NTC, platinum resistance thermometers exhibit excellent linearity and long-term resistance stability over a wider operating temperature range. Deploying them in the diffusion region, which is responsible for heat conduction, allows for stable and reliable monitoring of continuous temperature gradient changes, providing high-precision steady-state data support for evaluating the module's long-term heat dissipation capability.

[0115] Furthermore, the system configures the temperature sensing element in the weakly interactive mesh unit as a high-precision digital temperature sensor. Since this area is located at the edge of the module, temperature changes are relatively weak and highly susceptible to interference from ambient electromagnetic signals. The digital temperature sensor, employing a built-in analog-to-digital converter chip, not only provides extremely high temperature resolution to identify minute heat dissipation anomalies but also directly outputs a digital signal with strong anti-interference capabilities, effectively avoiding the attenuation and distortion of weak analog signals during long-distance transmission.

[0116] After clarifying the component selection for each node, this embodiment further specifies that each sensing node in the digital sensing array is integrated and arranged on a flexible printed circuit board (FPC) for the physical form and wiring architecture of the entire sensing array. This integrated design brings significant technical advantages in multiple dimensions: On the one hand, the flexible printed circuit board has excellent bending and conformal characteristics, which can better fit the complex three-dimensional surface morphology of the module body, such as the undulation of the top cover of the battery cell or the curvature of the side plate, thereby ensuring reliable contact between the bottom heat conduction structure and the module surface; on the other hand, the use of FPC integrated wiring effectively avoids the problems of large space occupation, easy wear of wire harnesses and poor assembly consistency caused by traditional discrete independent wiring. More importantly, by etching and fixing all nodes on the same flexible printed circuit board, the relative spatial coordinates between each dual-parameter sensing node are locked in the physical structure. This high-precision physical position fixation provides relatively stable and accurate geometric spatial prior data for subsequent non-uniform spatial interpolation processing of continuous three-dimensional thermal field models based on spatial topological relationships.

[0117] Any aspects of this invention not described in detail are well-known to those skilled in the art.

[0118] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for sensing and analyzing the thermal field of a module, applied to data processing equipment, characterized in that, include: The module acquires real-time operating power and acquires temperature and heat flow data of the module grid cells collected by a digital sensing array deployed on the surface of the module body via a heat-conducting structure. The digital sensing array has non-uniform sensing characteristics, and the module grid cells are pre-divided into core, diffusion, and weak interaction layer regions according to the heat dissipation intensity characteristics of the module. The non-uniform sensing characteristics include at least deploying sensing nodes with different performance parameters in the core, diffusion, and weak interaction layer regions. Based on the spatial topological relationship of the multiple hierarchical regions and the heat flux density variation trend reflected by the heat flux data, non-uniform spatial interpolation processing is performed on the temperature data and the heat flux data to construct a continuous three-dimensional thermal field model of the module. Extract the temperature distribution characteristics and heat flow distribution characteristics from the three-dimensional thermal field model, and calculate the core thermal characteristic quantitative indicators of the module. The core thermal characteristic quantitative indicators include at least the thermal field uniformity index calculated based on the temperature distribution difference measurement, the heat dissipation capacity index calculated based on the heat flow density, and the heat accumulation risk index calculated by combining the extreme values ​​of temperature and heat flow. Based on the aforementioned core thermal characteristic quantification indicators, the system outputs early warning results for the dynamic evolution of the thermal field of the module.

2. The module thermal field sensing and characteristic analysis method according to claim 1, characterized in that, The process of calculating the core thermal characteristic quantitative index includes: Calculate the standard deviation or variance of the real-time temperature value of each sensing node in the digital sensing array and the average temperature value of all sensing nodes, and construct the thermal field uniformity index based on the ratio of the standard deviation or variance to the average temperature. The average value of the real-time heat flow data acquired by the sensing nodes deployed in the diffusion grid unit is calculated, and the ratio of this average value to the real-time operating power of the module is calculated, which is used as the heat dissipation capacity index.

3. The module thermal field sensing and characteristic analysis method according to claim 1, characterized in that, The process of calculating the core thermal characteristic quantitative index includes: Determine the first ratio of the maximum heat flux data to the average heat flux data of all sensing nodes in the digital sensing array, and the second ratio of the maximum temperature data to the average temperature data. The first ratio and the second ratio are multiplied to obtain the heat accumulation risk index used to characterize the tendency of local overheating; The process of outputting early warning results of the dynamic evolution of the thermal field for the module based on the core thermal characteristic quantification index includes: The standardized temperature and heat flow digital signals after analog-to-digital conversion are transmitted to peripheral devices; and when the heat accumulation risk index is greater than or equal to a preset risk threshold, a local overheating warning signal is generated.

4. A module thermal field sensing and characteristic analysis system, characterized in that, Includes a data processing device and a digital sensing array communicatively connected to the data processing device; The digital sensing array is attached to the surface of the module body through a heat-conducting structure and is divided into multiple hierarchical regions according to the heat dissipation intensity characteristics of the module body. The multiple hierarchical regions include core grid units with dense heat flow, diffusion grid units with diffused heat flow, and weak interaction grid units with weak heat flow interaction. The digital sensing array includes dual-parameter sensing nodes deployed in the multiple hierarchical regions, and each dual-parameter sensing node is equipped with a temperature sensing element and a heat flow sensing element. The data processing device includes a processor and a memory, the memory being used to store a computer program, and the processor executing the computer program to implement the module thermal field sensing and characteristic analysis method as described in any one of claims 1 to 3.

5. The module thermal field sensing and characteristic analysis system according to claim 4, characterized in that, To adapt to the heat flow characteristics of each level of region: The response speed of the temperature sensing element deployed in the core grid cell is greater than that of the temperature sensing element deployed in the diffusion grid cell, in order to capture transient heat flow accumulation; The temperature sensing elements deployed in the weakly interactive grid cells have a higher measurement resolution than those deployed in the diffused grid cells, enabling them to accurately capture boundary heat dissipation anomalies and minute temperature fluctuations.

6. The module thermal field sensing and characteristic analysis system according to claim 4, characterized in that, The temperature sensing element and the heat flow sensing element in the dual-parameter sensing node are arranged side by side in space, and a heat insulation groove is provided between the temperature sensing element and the heat flow sensing element. The dual-parameter sensing node is covered with an insulating structure on top and has a heat-conducting structure connected to the module body on the bottom. The combination of lateral heat transfer and heat conduction is blocked by the upper insulation and the middle heat-insulating groove, so as to isolate the interference of the temperature sensing element's own heating on the heat flow sensing element's measurement.

7. The module thermal field sensing and characteristic analysis system according to claim 6, characterized in that, The thermally conductive structure is a gradient thermally conductive bonding structure, which is configured with a stepped thermal conductivity based on the heat dissipation intensity characteristics of the multiple layer regions.

8. The module thermal field sensing and characteristic analysis system according to claim 7, characterized in that, The thermal conductivity of the thermally conductive structure in the region corresponding to the core grid unit is greater than that in the region corresponding to the diffuse grid unit, and the thermal conductivity of the thermally conductive structure in the region corresponding to the diffuse grid unit is greater than that in the region corresponding to the weakly interactive grid unit.

9. The module thermal field sensing and characteristic analysis system according to claim 8, characterized in that, The gradient thermally conductive bonding structure achieves a step-like change in thermal conductivity by setting thermally conductive silicone with different thicknesses and intrinsic thermal conductivity in different layer regions.

10. The module thermal field sensing and characteristic analysis system according to claim 4, characterized in that, The temperature sensing element in the core grid unit is an NTC thermistor, the temperature sensing element in the diffused grid unit is a platinum resistance thermometer, and the temperature sensing element in the weakly interactive grid unit is a high-precision digital temperature sensor; each sensing node in the digital sensing array is integrated and arranged on a flexible circuit board.

Images